Endothelin-1 decreases microvascular fluid leak via second messenger cyclic nucleotides and protein kinase a signal transduction

276

ASSOCIATION FOR ACADEMIC SURGERY—ABSTRACTS

10. Hypoxic Pulmonary Vasoconstriction and Pulmonary Artery Tissue Cytokine Expression Are Mediated by Protein Kinase C. B.M. Tsai, MD, M. Wang, MD, J.M. Pitcher, MD, and D.R. Meldrum, MD. Indiana University School of Medicine. Background. Pulmonary arteries exhibit a marked vasoconstriction when exposed to hypoxic conditions. Although this may be an adaptive response to match lung ventilation with perfusion, the potential consequences of sustained pulmonary vasoconstriction include pulmonary hypertension and right heart failure. The concomitant production of proinflammatory mediators by the pulmonary artery itself may exacerbate acute increases in pulmonary vascular resistance. We hypothesized that acute hypoxia causes pulmonary arterial contraction and increases the pulmonary artery tissue expression of proinflammatory cytokines via a protein kinase C (PKC) mediated mechanism. Methods. Isometric force displacement was measured in isolated rat pulmonary artery rings during hypoxia (95% N 2/5% CO 2, pO 2 ⫽ 30 –35 mmHg) in the presence and absence of the protein kinase C inhibitor chelerythrine (1 ␮mol/L). Following 60 min of hypoxia, pulmonary artery rings were subjected to mRNA analysis for TNF-␮, IL-1␤, and iNOS via RT-PCR. Data were analyzed using two-way analysis of variance (ANOVA) with post-hoc Bonferonni test or unpaired t tests with alpha level less than 0.05 considered statistically significant. Results. Hypoxia caused a biphasic contraction: an early and delayed contraction which occurred 1–3 and 15–20 min, respectively, after the onset of hypoxia. Hypoxic pulmonary artery tissue had increased expression of TNF-␮, IL-1␤, and iNOS mRNA compared to normoxic controls. PKC inhibition significantly (P ⬍ 0.001) attenuated delayed hypoxic contraction (61.55 ⫾ 3.91% versus 94.07 ⫾ 5.94% in hypoxia alone) and prevented hypoxia-induced pulmonary artery tissue expression of TNF-␮, IL-1␤, and iNOS mRNA. Conclusions. These findings demonstrate that hypoxia results in pulmonary artery contraction and promotes the expression of inflammatory mediators. Both processes are mediated by PKC. We conclude that there may be a therapeutic role for PKC inhibition in the treatment of acute hypoxic pulmonary vasoconstriction.

creased levels of cGMP alone, ET-1 in the presence of increased cGMP levels decreased L p from 2.3 ⫾ 0.2 to 1.5 ⫾ 0.1 (P ⬍ 0.008, n ⫽ 6). Conclusion. The permeability-decreasing effect of ET-1 was blocked by cAMP and PKA inhibition. In contrast, ET-1 was able to decrease fluid leak while cGMP degradation was inhibited. The intracellular mechanism of ET-1 may involve increased cAMP production and PKA activation, but not cGMP degradation. Further understanding of intracellular mechanisms that control microvascular fluid leak may lead to the development of a pharmacologic therapy to control third space fluid loss in severely injured or septic patients.

12. Chylomicron-Bound LPS Inhibits Tumor Necrosis FactorInduced Capillary Leak In Vivo. A.L. Spitzer, MD, G.P. Victorino, MD, F. Kasravi, MD, PhD, B. Curran, BS, J. Chang, BS, and H.W. Harris, MD, MPH. University of California, San Francisco and East Bay, San Francisco and Oakland, CA. TNF plays a central role in the host acute inflammatory response during sepsis. A potent mediator of inflammation, this cytokine has been shown to increase microvascular permeability through a direct effect on endothelial cells. Previously, we have demonstrated that chylomicron (CM)-bound LPS inhibits the hepatocellular response to TNF-induced stimulation. Therefore, we hypothesized that CM-LPS can also protect endothelial cells from the deleterious effects of TNF. Sprague-Dawley rats were injected with saline (control), CM (375 mg/kg BW), LPS (19 ␮g/kg BW), or combined CM-LPS complexes, 4 h prior to TNF exposure. We then examined the effect of TNF (1 ␮g/mL) applied intraluminally on microvascular permeability (hydraulic conductivity) in post-capillary venules using the Landis technique. As expected, TNF stimulation increased vascular permeability approximately 3-fold as compared to baseline (Fig. 1). CM alone

11. Endothelin-1 Decreases Microvascular Fluid Leak Via Second Messenger Cyclic Nucleotides and Protein Kinase a Signal Transduction. T.J. Chong, MD, B. Curran, BS, G.P. Victorino, MD. University of California, San Francisco--East Bay. Introduction. Endothelin-1 (ET-1) decreases microvascular fluid leak by an unknown mechanism. Elevated cAMP levels, depressed cGMP levels, and protein kinase A (PKA) activation are all known to decrease fluid leak. We hypothesized that ET-1 decreases permeability by increased cAMP levels, decreased cGMP levels, and PKA activation. The purpose of this series of experiments was to determine ET-1’s effect on venular fluid leak during (1) cAMP synthesis inhibition, (2) increased cGMP levels, and (3) PKA activation inhibition. Methods. Using the modified-Landis technique, rat mesenteric venules were cannulated to measure hydraulic permeability, L p (units ⫻ 10 ⫺7 cm/s/cmH 2O). L p was measured during continuous perfusion of ET-1 and a test solution. The test solutions consisted of (1) a cAMP synthesis inhibitor (2⬘,5⬘ddA), (2) an inhibitor of cGMP degradation (zaprinast), or (3) an inhibitor of PKA (H-89). These results were compared to L p measurements from the different test solutions alone. Data were statistically analyzed using unpaired t-tests. L p values are represented as mean ⫾ standard error. Results. ET-1 did not change L p during (1) cAMP synthesis inhibition and (2) PKA inhibition: cAMP inhibitor with and without ET-1 (L p ⫽ 3.6 ⫾ 0.1 versus 3.6 ⫾ 0.2, P ⫽ 0.4, n ⫽ 4) and PKA inhibitor with and without ET-1 (L p ⫽ 2.3 ⫾ 0.2 versus 2.6 ⫾ 0.2, P ⫽ 0.3, n ⫽ 5). Increased cGMP levels (inhibition of cGMP degradation) failed to block the permeability-decreasing effect of ET-1. Compared to in-

Figure 1 Rat microvascular permeability in response to TNF. Data represent the mean ⫾ SEM of ⱖ3 separate experiments (*P ⬍ 0.001, t-test compared with other subsets).

partially inhibited the TNF-induced increased permeability. However, pretreatment with CM-LPS completely abolished the TNFinduced changes in vascular permeability. Conversely, the animals injected with LPS alone yielded an exaggerated response to TNF. CM-LPS inhibited TNF-induced increases in vascular permeability in vivo. Consistent with our previous observations in hepatocytes, our data suggest that lipoprotein-bound LPS can similarly induce cytokine tolerance in vascular endothelial cells, yielding further support for the development of a lipid-based therapy for gram negative sepsis.